Market Vector Auto Regression High CPU 1.2
Device: iOS iPhone
Category: Finance
Price: $49.99 -> Free, Version: 1.2 (iTunes)
Description:- New dedicated compute cluster for High CPU edition, powered by Intel Xeon chips
- More backtests allowed per trial (up to 500 trading days) for High CPU edition
This app models the (close-open) price direction (that is, today's price travel) of a given target stock as a function of prior days opening and closing prices of related stocks. Also works for ETFs if they are available in Yahoo Finance historical data.
Some use cases:
- You want to keep a position open for 1 trading day only. Models built with this app can give an estimate as to whether the stock/ETF will go up or down today.
- You want to buy or sell a given stock/ETF for other reasons. Models built with this app can give an estimate whether Today is a good day to buy or sell, or whether you might want to wait until a later day to make a trade.
No forecast is 100% accurate. Estimated accuracy of forecasts is computed by backtesting. Whether you should rely on long backtest probabilities or short backtest probabilities is a judgement call on your part.
Some thoughts on backtesting, to assist with judgement calls:
http://ift.tt/1letHtq
From Wikipedia:
"The only prior knowledge required is a list of variables which can be hypothesized to affect each other intertemporally."
http://ift.tt/pEGMXo
For example, one might think that the price change of General Motors today (GM) might depend on the recent prices of oil (USO is an oil ETF). This can be modeled using a tool such as this.
Note that there may not be a predictive relationship for your chosen symbols. In this case, a model's backtest will be poor (low % correct) and the model is not useful for forecasting.
In the case where a backtest yields reasonable results, the model may have some predictive
power for 1 day ahead forecasts.
Only price travel Direction is attempted to be modeled, not the actual price change in dollars.
Quickstart:
1. enter symbol to forecast and candidate predictor symbols.
2. select Backtest or Forecast.
3. Press Run. Calculation results will appear in green window.
Further details at http://ift.tt/PaIBnh
Features:
Up to 4 candidate predictor stocks/ETFs (including target).
3 model types generated for every forecast for comparison purposes:
annealing classifier (slow or fast)
linear least squares
k-nearest classifier
1 and 2 day data lags automatically generated [AR(2) type models]
Up to 500 trading days of backtests.
1 day forward forecast.
Model can be built/forecasted before market opens since "today's" open prices are not included in the model.
Public data feed ; no feed fees.
Detailed log file of calculations for auditing.
Calculations done on a server for battery conservation.
Ability to Stop long calculations on the server.
Downloadable assembled regression tables with dates aligned in CSV format for further study:
1. raw prices table
2. daily deltas table
3. daily deltas with future "known" X vector computed for you
Ability to send downloaded tables to other apps on your device or via email.
No user email or account setup required.
Simultanenous runs on multiple iOS devices are allowed if devices are not sync'd.
Common operation among devices if devices are sync'd.
What's New
Add backtest metrics:
1. Report out maximum and average (mean) bad run length per model. This measures the maximum sequence of bad predictions that occurs during the backtest.
1a. The mean is the mean of all bad run sequences during the backtest. "Good" run sequences, where we have a sequence of correct predictions, are not included in the mean calculation.
2. Add verbal notation to backtest probabilities:
Poor: probability of achieving results randomly is more than 20%
Marginal: probability of achieving results randomly is between 10% and 20%
Fair: probability of achieving results randomly is between 5% and 10%
Good: probability of achieving results randomly is between 1% and 5%
Very Good: probability of achieving results randomly is less than 1%
These probabilities are computed from the overall % correct number of the backtest, not the max or mean bad run lengths.
3. Speed up Annealing model generation
4. Set the random number seed properly so that repeated trials on the same data will yield the same Annealing models (model 1).
5. Plot tab:
Add plots of the bad prediction runs that occurred during the backtest.
6. Tune tab:
Add new features to allow user-controlled manual tuning of the regression window size and the K in the K-nearest method. By increasing or decreasing these values, you may be able to improve the model backtest quality. Improved model backtests may suggest improved forecast accuracy. Note that the ideal model tuning setup will likely be different, for a different number of backtests and for different target and predictor stocks.
Market Vector Auto Regression High CPU
0 comments:
Post a Comment